Thongsy
3941
Poor Dan Quayle, just gets no respect. Is there a reason why they took that out? I loved seeing how I did throughout the game, when I was excelling and when I wasn’t.
He’s still on the end-game screen. Was there ever an in-game comparison list?
It’s still there. you just have to click a tab on the endgame screen.
hong
3944
… okay, I can officially state that the Civ 5 interface sucks.
Because the extremely obvious button to select that list is only gigantic rather than covering the entire screen?
Quitch
3946
I like that people have made it 2010 without encountering tabbed score screens before.
The AI seems to think that ranged units are melee in general. I can’t count the number of times I’ve seen it move archers / artillery adjacent to a target that it could have fired on from a distance. Which 95% of the time gets the unit killed.
They haven’t researched it yet.
Quiet you. It’s been officially declared. Stop questioning the authorities.
jpinard
3950
City - City specialization and city focus improvements.
Anyone know exactly what this means? In my last game I didn’t really see anything that made city specialization any better.
There’s a good base to build on in Civ V but so much I disagree with. I’ve logged over 100 hours and more-so than ever feel the removal of the health game-play mechanic was a mistake. Of course this led to City-States which being just generic options was a bad direction to go. Though it may be realistic to have a city show up in the middle of the desert - which can technically get food via trade, it just takes away the fun of trying to find super city locations. I really loved that in the previous Civ’s.
And the loss of end-game replays & historical graphs. I’m like Gollum constantly saying, “My precious, where is my precious?!” - hehe
Jf1513
3951
It didn’t list anything saying they implemented any performance improvements, but the game does seem like it runs a bit better, and reading some forums (such as Civfanatics) it seems like some people are getting better performance as well. I didn’t really have long times in between turns beforehand, and now it seems somewhat faster than before, which is always nice.
The “city focus improvements” probably refers to better tile selection when you choose Gold or Production for the automated citizen placer. It does seem slightly less stupid, in that I only found myself correcting it on placement a time or two. One time it absolutely wanted to send Paris into starvation. Every time something reset the city, like a tech advance, bang Paris was in starvation again.
Puppets are significantly better now. They focus very heavily on gold, and by and large don’t build crap I’d never touch like the stables. Now they’re a net gain all the way around. If you’re running 5-6 puppets you’ll be swimming in money, though they tend to be very poor at production. I now annex puppets only if I have something specific I want to do with the city, like build wonders or military units.
The AI, however, is still very sad. It simply does not handle naval units well at all - they tend to go completely passive, not bothering to fire back at the ships that are slowly eliminating them. Now and then they’ll fight back, but all too often they do nothing when faced with my fleet. I’ve seen some sizeable fleets too, like one with 3 battleships.
Enemy cavalry seems to show up now and then even if the civ doesn’t have a unique cavalry unit, but I still haven’t seen an air force. The most I’ve seen in the way of air defenses in the occasional AA gun. This means I typically overrun the second continent with mech infantry and bombers or stealth bombers.
It doesn’t seem like it should be too hard to fix things like “not firing when the opportunity is there.”
I’m starting to wonder if the cavalry/AA guns that people occasionally see under AI control aren’t just city state gifts. . . city states seem to fucking LOVE giving me AA guns despite me being the only person in the world with the power of flight.
Dejin
3955
In my game against Germany mentioned previously, where Germany refused to build capital ships, but choked its continent almost completely full of ground units, it seemed to have a fair bit of anti-aircraft. It started with standard anti-air units and then started sending out mobile SAMs. I think there were far too many to be all city state gifts, plus most of the city states didn’t like Germany. This was all pre-patch.
In one of my Emperor difficulty games, the Siamese would fill the continent with mobile sams and mechanized inf in a ratio of 2:1. Additionally they filled a small pond in the middle of the continent with about 10 destroyers but didn’t have a single ship in the actual ocean that surrounded our content. So they were open to naval invasions from everyone from any of the other continents… And they just loved moving their expencive rocket artilleries right in front of my modern armors and mech infs and in range of my own rocket artilleries. It was no fun dead-clicking hundreds of their units without them ever putting up a real fight. By 1900 I had 4 rocket artilleries at level 8 with +1 range and +1 attack per round. so they alone were litterally killing his units faster than he could produce them (at a rate of 4-6 per turn). With such stupidity even larger production buffs wouldn’t help the AI, unfortunately.
That’s been a problem with Civ AI since at least Civ 2.
Now, why don’t they put up a prize, such as $10.000, provide a well-defined interface to the game and let some people out there program the AI (or certain parts of it) by random people on the 'net in a competition? Not only would it provide several differently playing AIs and better AIs in general, it would also generate some additional marketing…
I know my AI wouldn’t fill useless ponds with destroyers.
Jon Shafer already said that the SDK including AI source code will be released, just as for Civ4, and plenty of people will try to improve it even without a prize.
Yes, but putting up a competition before/very shortly after the game releases, and integrating the results into the game properly produces an entirely different result than an unconcentrated effort by modders which may end up with something decent a year after release at the earliest, and only if you know which of the by then many mods will really improve things. And this also without a proper comparison that is only possible through a proper competition (like the current Galcon one at google).
I actually believe that after sound engines, 3D engines, speed tree engines and the like, also something like a AI engine is needed, that can be lizenced and would consist of a set of good high-level strategizing concepts, perfect way-finding algorithms for the various scenarios (exact, fuzzy, masses of units, puzzle-like scenarios, etc.) and more. Except game-companies currently probably still didn’t get enough flak for bad AI to consider shelling out significant money for such a thing.
